JP2794472B2 - Rotary weighing method - Google Patents

Description

Description: FIELD OF THE INVENTION The present invention relates to a weighing method using a rotary weighing device in which a rotating means rotates around a center of rotation, and more particularly to a weighing method generated in connection with centrifugal force generated by rotation. To correct errors that occur.

[Prior Art] Conventionally, in a rotary measuring device, an article 4 is loaded on a measuring means 2 such as a load cell coupled to a rotation center O as shown in FIG. Something to measure. In such a rotary metering device, if the metering means 2 by the weight of the article is rotated deflected by an angle theta, it is required the force F ₁ which balances the centrifugal force F. This force F ₁
Occur as horizontal component of force F ₂ generated in parallel to the horizontal beams 8 of the metering means 2. Such force F ₂ is, F ₁ perpendicular force thereto, so that is, the resultant force of the vertical component force F _3, is the vertical component force F ₃ is applied upward to the metering means 2, metering means 2 Accurate weighing was hindered. The vertical component force F ₃ is the size of F _tan theta As is clear from Figure 12. Also, F
Is the distance between the rotation center O and the measuring means 2 as r,
F = mv ² / r, where m is the total mass of the measuring device 2 and the rotating speed of the rotary measuring device is v.

In order to eliminate the effect of such F _3, Sho 63-55
Japanese Patent Application Publication No. 002 discloses that the above-mentioned F ₃ is calculated by F _tan θ and added to the output of the measuring means 2.

[Problems to be Solved] However, in the above publication merely seeking F _3, it is only described to be added to the output of the metering means 2,
Not described in detail for how a removing F _3.

[Means for Solving the Problems] A first method of the present invention is directed to a rotary type weighing device having weighing means that rotates at a constant speed around a predetermined center of rotation, in an initial state in which no articles are loaded on the weighing means. Measuring the static variation output of the weighing means when the rotary weighing device is in a non-rotating state, and in the initial state, the rotary weighing device is rotating at a predetermined speed. Initial centrifugal force error, which is the vertical component of the force generated to balance the centrifugal force generated during rotation of the rotary measuring device in the initial state, based on the output of the weighing means in the state and the static fluctuation amount. Measuring, based on the output of the weighing means and the static variation when the rotary weighing device is rotating at a predetermined speed while the reference article is loaded on the weighing means, To the loading condition of reference goods Obtaining a centrifugal force error when the reference article is loaded, which is a vertical component force of the force generated to balance the centrifugal force generated at the time of rotation in;
Determining whether or not an article to be weighed having a weight close to the weight of the reference article is loaded in the rotating state of the rotary weighing device at a predetermined speed; and determining that the article to be weighed is loaded. Calculating the weight of the article to be weighed based on the output of the weighing means, the stationary fluctuation, and the centrifugal force error when the reference article is loaded; and determining that the article to be weighed is not loaded, And correcting the static fluctuation on the basis of the output of the above and the initial centrifugal force error.

The second invention is a step of measuring an initial value variation which is an output of the weighing means when the rotary weighing apparatus is in a non-rotating state in an initial state where no articles are loaded on the weighing means; The step of determining the true static weighing value of the means itself, the output of the weighing means and the rotational speed and the initial value of the rotary weighing apparatus in the measuring state where the articles to be weighed are loaded on the weighing means and the rotary weighing apparatus is rotating The centrifugal force error, which is the vertical component of the force generated to balance the centrifugal force generated during rotation of the rotary weighing device in the loaded state of the articles to be weighed based on the fluctuation component and the true static weighing value of the weighing means, is used. Calculating the weight of the article to be weighed based on the output of the weighing means in the measurement state and the calculated centrifugal force error.

A third invention is a step of measuring a static variation which is an output of the weighing means when the rotary weighing apparatus is in a non-rotational state in an initial state where no articles are loaded on the weighing means; The vertical force of the force generated to balance the centrifugal force generated based on the rotation of the reference article loaded on the weighing means based on the weight of the weighing device, the rotation speed of the rotary weighing device, and the initial load of the weighing means. Calculating a centrifugal force error when the reference article is loaded, which is a component force; and outputting and stopping the weighing means in a rotating state of the rotary weighing apparatus when an article to be weighed having a weight near the weight of the reference article is loaded on the weighing means. Calculating the weight of the article to be weighed based on the variation and the centrifugal force error when the reference article is loaded.

[Operation] According to the first invention, the centrifugal force error generated when the weighing unit is rotated while the reference article is loaded is measured in advance, and the weighed article having a weight close to the weight of the reference article is measured. The output of the weighing means when the weighing means is rotated with the article loaded thereon includes a centrifugal force error substantially equal to the centrifugal force error when the reference article is loaded, and an initial stationary fluctuation. Therefore, a true weighing value of the article to be weighed is obtained based on the output of the weighing means, the centrifugal force error, and the initial fluctuation. Further, the initial variation changes with time. Therefore, the initial static fluctuation is corrected based on the output of the weighing unit when the article to be weighed is not loaded and the initial centrifugal force error. That is, the output of the weighing means when the article to be weighed is not loaded is composed of the initial stationary fluctuation and the initial centrifugal force error, and the initial centrifugal force error is a constant value. Therefore, by calculating the output of the weighing means when the article to be weighed is not loaded and the initial centrifugal force error, the initial fluctuation can be updated to the current value.

The second invention is to remove the initial stationary fluctuation from the output of the measuring means when the rotary measuring device is rotating, and to obtain a centrifugal force error based on this and the rotation speed. Although the initial stationary fluctuation includes the measured value of the measuring means and the fluctuation of the zero point, there is a possibility that an error has already been included in a state where these are measured. Therefore, the true weighing value of the weighing means itself is separately obtained, and the weight of the article to be weighed and the weighing means itself is obtained based on the true weighing value of the weighing means itself, the output of the weighing means, and the initial variation. The centrifugal force error is determined based on this and the rotation speed. Strictly speaking,
The weight between the article to be weighed and the weighing means itself includes a centrifugal force error, but since the value is small, there is no significant effect.

In the third invention, a centrifugal force error at the time of loading the reference article is calculated based on the weight of the reference article, the initial variation, and the weight of the weighing means itself, and the true weight of the article to be measured is measured using this. doing.

Embodiment FIG. 1 to FIG. 9 show a first embodiment. In this embodiment, the weight of an article to be weighed is close to a reference weight M, such as a weight sorter. The present invention is applied to a weighing apparatus 10.

As shown in FIG. 7, the rotary measuring device 10 includes n measuring means provided to rotate around a rotation center O.
It has 12 _{1 to} 12 _n . These measuring means 12 _{1 to} 12
_n it is weighed and the platform 14 ₁ to 14 _n, the load cell 16 ₁ to 16 coupled respectively thereto _n, has. These metering means 12 ₁ to 12 _n is configured to be rotated at a predetermined constant speed in the direction of arrow A by a driving source not shown. These weighing means 12 ₁ to 12 _n, un weighing of the article 20 is sequentially fed by the star wheel 18, metering completion of article 20 conveyor 24 by the star wheel 22
To be carried out. The unweighed articles 20 supplied to the star wheel 18 are also transported by the transport conveyor 24,
A screw conveyor 26 is provided to feed unweighed articles 20 in synchronization with the rotation of the star wheel 18.
28 is a guide. The rotations of the star wheel 18, the rotary weighing device 10, and the star wheel 22 are also synchronized so that the delivery of the articles 20 is performed accurately.

Analog weight signal which is an output of each load cell 16 ₁ to 16 _n as shown in FIG. 8 is amplified by the amplifier 30 ₁ to 30 _n, via the analog switches 32 ₁ to 32 _n A / D
It is supplied to a converter 34, where it is converted into a digital weighing signal, which is supplied to a CPU 38 via an interface 36. The CPU 38 operates according to the program stored in the memory 40, and outputs pulse signals Ta and R from the pulse generator 42.
By controlling the analog switches 32 _{1 to} 32 _n and the A / D converter 34 based on command signals from the P, TP and the keyboard 44, the output signals from the load cells 16 _{1 to} 16 _n are digitized and input. And measuring means 12 _{1 to} 12 _n
Weight (initial load), eliminating the fluctuation of the zero point, further removing the centrifugal force error, obtaining the true weighed value of the article 20, comparing this with the reference weight, mainly sorting Do. The results of these selections are displayed on the display 46.

The outline of the calculation performed by the CPU 38 to obtain the true weighing value of the article 20 is as follows. For example, while the metering means 12 ₁ is rotated, assuming that by measuring the weight of the article 20, the output W of the load cell 16 ₁ at this time becomes _{W = W true + W tare +} W z -W error. However W _true true weight of the article 20, W _tare weight of the weighing means 12 _1, W _z is the zero point fluctuation, W _error is centrifugal force generated on the basis of the weight of the article 20, the weight of the weighing means 12 ₁ And the vertical component of the force generated to balance the load (referred to as centrifugal force error). It should be noted that W _error is negative because it acts upward as is apparent from FIG. To find W _true from the above equation, subtract W _tare and W _z from W
_Error can be added. And W _true and W _tare Therefore, metering means 12 ₁ is stationary, and the article 20 is weighed by the non-loading state, stores these metric values (referred to as initial value variation). Similarly, it is desired to store W _error , but this value fluctuates strictly according to the weight of the article 20. However, as described above, in this embodiment, since the article 20 is only one having a weight in the vicinity of the reference weight M, the centrifugal force error when an article having a basis weight M and loading the weighing means 12 ₁ Even if used as W _error , no large error occurs. Therefore, the centrifugal force error when an article having a basis weight M and loading the weighing means 12 ₁ (reference article loading Centrifugal force error)
Remember W _error '. Then, W _true + W _z is subtracted from W, and W _error ′ is added to obtain W _true .

However, W _z of W _tare + W _z varies with time change. Therefore, when the article 20 to the metering unit 12 ₁ is not loading, it is necessary to make the so-called zero-point adjustment. Goods
20 is not loading, the output W _o of the load cell 12 ₁ when rotating, becomes ". _{However, W error" W o = W} tare + W z -W error is the article 20 to the metering unit 12 This is the centrifugal force error that occurs when no load is applied. Therefore, W _o
By subtracting W _tare from, and adding W _error ″, a new zero point variation is obtained, and this is used to calculate
Ask for the true weight w _true . Therefore, in a state where the weighing means 12 is stationary and the article 20 is not loaded, W
_Tare and W _error ″ are measured and stored.

In order to perform the above-described processing, the CPU 38 initializes when the power is turned on as shown in FIG. 4, gives a static weighing command, and sets the static weighing flag SWF to 1 (step S1). S2). Then, the measurement end flag WEGF is set to 1
Is determined (step S4), and this step S4 is YES
Step S4 is repeated until. In this case, the metering means 12 ₁ to 12 _n is stopped, and the article 20 is non-loading.

On the other hand, the CPU 38 is interrupted every time the pulse generator 42 generates the pulse signal Ta, and the processing shown in FIGS. 2 and 3 is performed. That is, when an interrupt occurs as shown in FIG. 2, it is determined whether a dynamic weighing command has been given (step S6). However, since this answer is NO, as shown in FIG. In step S8, the timing pulse count control flag TPF used is set to 0 (step S8), and it is determined whether SWF is 1 (step S10). Since the SWF is set to 1 first, the answer to step S10 is YES. Next, it is determined whether or not the flag SSF which is set to 1 when the load cell to be weighed is specified is 1 (step S12). If the answer is NO, the flag SSF is set to 1 (step S1).
4), it is determined whether the flag RSF, which is set to 1 when the designation of each of the load cells 16 _{1 to} 16 _n is started, is 1 (step S).
16). If the answer is NO, the flag RSF and 1 (step S18), and the value of the counter SSC for specifying the analog switches 32 ₁ to 32 _n and thus the load cell 16 ₁ to 16 _n is 1 (step S20) , The analog switch corresponding to the value of the counter SSC (in this case analog switch 32 ₁ )
Is closed (step S22). This allows load cell 1
Output of 6 ₁ is supplied to the A / D converter 34. Subsequent to step S22, the flag WAITF, which is set to 1 when the analog switch is closed, is set to 1, and the flag WEIGF, which is set to 1 when the A / D conversion timing comes, is set to 0 (step S24). End the load routine.

Next, when the pulse signal Ta is generated and the interrupt routine is executed, step S12 is executed through steps S6, S8, S10. Since the answer is YES, the flag WEIGF is set to 0.
Is determined (step S26). This answer is NO because the flag WEIGF is set to 0 first, and then the flag WAIGF
It is determined whether TF is 1 (step S28). This answer is YES because the flag WAITF was previously set to 1, and the value of the time measurement counter WC is incremented by 1 (step S30).
It is determined whether the value of the counter WC is equal to a predetermined q1 (step S32), and if this answer is NO, this interrupt routine is ended. Thereafter, each time the pulse signal Ta is generated, steps S6, 8, 10, 12, 26, 28, 30, and 32 are repeated until the answer to step S32 is YES. When the answer to step S32 is YES, the value of the counter WC is set to 0, the flag WEIGF is set to 1, the flag WAITF is set to 0 (step S34), and the interrupt routine is ended. The reason for keeping the analog switch closed for a certain period of time is that if the A / D conversion is performed immediately after the analog switch is closed, the analog switch is connected to the output of the load cell supplied to the A / D converter 34. This is because the output of the load cell cannot be accurately A / D-converted because it includes noise and the like generated when closing.

Next, when an interrupt is made, step S26 is performed in the same manner as described above.
The answer to step S26 is that the flag WEIGF
Is set to 1, so the answer is YES. Therefore, the A / D converter 34
Giving A / D conversion command signal, converts the output of the load cell 16 ₁ to the digital weight signals Wi, Reads a this (step
S36). Note that the A / D converter 34 can terminate the A / D conversion between the generation of the pulse signal Ta and the generation of the next pulse signal Ta. The value of the counter WEC which is accumulated in the read digital weighing signal Wi accumulation register $ Wi and counts the accumulated number is incremented by 1 (steps S38, S40). Then, it is determined whether the value of the counter WEC is equal to a predetermined number m1 (step S42). If the answer is NO, the interrupt routine is terminated. Hereinafter, similarly, each time the pulse signal Ta is generated, the digital weighing signal Wi is read, accumulated, and counted. When the count value becomes equal to m1 (the answer in step S42 is YES), the register #Wi Is transferred to the register WINT, the value of the counter SSC is transferred to the register N (step S44), and the flag WEGF is set to 1
The register $ Wi, the counter WEC, and the flag SSF are set to 0 (step S46), and the interrupt routine ends.

Since the flag WEGF has thus become 1, the answer to step S4 shown in FIG. 4 is YES. Therefore, register WINT
The stored value is divided by m1, calculates the WI (weight in this case metering means 12 ₁₎ (step S48). And the flag WEG
F is set to 0 (step S50), and the calculated WI is stored in the register N
Is set in the initial load register specified by the value of (step S52). Then, the value of the counter CAC that counts the number of the weighing means for which the initial load is set is increased by one (step S54), and it is determined whether or not the value is equal to the number n of all the weighing means (step S56). , The process returns to step S4. Hereinafter, the initial load of other weighing means is set similarly. Therefore, in FIG. 3, if the answer to step S16 is YES, that is, if the flag RSF is 1, the value of the counter SSC is advanced by 1 to close the next analog switch (step S58), and step S22 is executed. I do. When the initial load setting of all the weighing means is completed in this way,
The answer to step S56 is YES, the value of the counter CAC is set to 0, the flag RSF is set to 0 (step S60), and the power-on routine ends. If the answer in step S10 is NO
, The register $ Wi and the counters WEC, WC, and SSC are each set to 0 (step S62), the flags SSF and RSF are set to 0, and the SWF is set to 1 (step S64).

Thus after finishing a set of initial load of the metering means 12 ₁ to 12 _n, when giving an indication of the zero point adjustment in a static state from the keyboard 44, the zero adjustment process routine shown in FIG. 5 (b) is performed You. At this time, each time the pulse signal Ta is generated, the interrupt routine shown in FIG. 3 is executed, and each time the weighing end flag WEGF becomes 1, the value of the register N is indicated in the register WINT. Output of the load cell (the value obtained by adding the initial load of the weighing means and the zero point fluctuation)
The value accumulated by m1 is stored.

In this routine, it is first determined whether or not the flag WEGF is 1 (step S66), and if this answer is NO, this routine ends. If the answer to step S66 is YES, the stored value of the register WINT at that time is divided by m1, and the average value W _{GSN of the} value obtained by adding the initial load of the weighing means specified by the register N and the zero-point variation is obtained. Is obtained (step S68). Then, the weighing end flag WEGF to 0 (step S70), the stored value W _IN of initial load register W which register N specifies
_By subtracting from the _GSN , a zero-point fluctuation _WZN is calculated (step S7).
2) This zero-point variation _WZN is stored in the zero-point register specified by the register N (step S74), and this routine ends. Hereinafter, as long as the zero point adjustment instruction is given, the zero point fluctuation amount for each weighing means is sequentially stored.

In the stationary state, when an instruction for normal processing is given from the keyboard 44, a normal processing routine shown in FIG. 5A is executed. In this case as well, each time the pulse signal Ta is generated, the portion shown in FIG. 3 of the interrupt routine is executed.
WINT indicates the output of the load cell indicated by the value of the register N (if the article 20 is not loaded on the weighing means, the value obtained by adding the initial load of the weighing means and the zero-point variation, and the article 20 is loaded on the weighing means. In this case, a value obtained by accumulating m1 values of the initial load of the weighing means, the zero-point variation, and the static weight of the article 20) is stored. In this routine, first, it is determined whether or not the flag WEGF is 1 (step S76). If the answer is NO, this routine is terminated. Further, when the answer to step S76 is YES, the value of the register WINT is divided by m1, seeking W _GSN (step S78), the flag WEGF 0
(Step S80). The _{WGSN at} this time is an average value of the initial load of the weighing means and the zero-point variation when the article 20 is not loaded on the weighing means, and when the article 20 is loaded on the weighing means, It is an average value of the initial load, the zero-point variation, and the stationary weight of the article 20. The W subtracted and indicated initial load W _IN and zeros variation W _ZN in register N from _GSN, calculates the W _SN (step S82). This W _SN
The value is 0 when the article 20 is not loaded on the weighing means, and is the static weight of the article 20 when the article 20 is loaded on the weighing means. Then, it is determined whether an instruction to display this _WSN is given from the keyboard 44 (step S8).
4), if the answer is NO, this routine ends.
If the answer to step S84 is YES, W _sn is displayed on the display 46, and this routine ends.

In order to remove the influence of the centrifugal force error from the weighing signal when the rotary weighing device 10 is rotating, the reference article centrifugal force error when an article having the reference weight M is loaded on each weighing means as described above. Must be calculated. Further, in order to correct the zero point fluctuation, it is necessary to calculate an initial centrifugal force error when no article is loaded. In any case, a weighing signal must be supplied from each load cell to the CPU 38 while each weighing means is rotating. Moreover, in that case, it is desirable that each weighing signal is stable. Therefore, in this embodiment, weighing is performed when each weighing means reaches a point a near the star wheel 22 shown in FIG. Therefore, the pulse generator 42 is configured to generate a pulse signal TP as shown in FIG. 9 each time each measuring means reaches the point a. Further, in order to metering means it has reached the point a is determined whether it is a metering means, a specific metering means, for example, each time 12 ₁ reaches the point a, the pulse generator 42 generates a pulse signal RP It is also configured as follows. Using the pulse signal TP and the pulse signal RP, a value obtained by accumulating the output of each measuring means by m by the portion shown in FIG. 2 in the interrupt by the pulse signal Ta is sequentially calculated.

That is, when an interrupt occurs, it is determined in step S6 whether a dynamic weighing command is given as shown in FIG. 2, and if the answer is YES, a flag S which is set to 1 at the time of static weighing is determined.
WF is set to 0 (step S62), and it is determined whether the flag TPF which is set to 1 when the pulse signal TP is supplied is 1 (step S64). If this answer is NO, the pulse signal TP
Is determined (step S66), and the answer is YES.
, The flag TPF is set to 1 (step S68), and it is determined whether or not the pulse signal RP is generated (step S7).
0). If this answer is YES, the weighing means has reached a point a since it is 12 _1, the value of the counter SSC to specify the analog switches turn each load cell 1 (step S72). Then, the analog switch designated by the counter SSC (in this case, the analog switch 32 ₁ ) is closed (step S74), the flag WAITF is set to 1, the WEIGF is set to 0, the counter WC is set to 0 (step S76), and the counter VC is set. The value is transferred to the register CAU, and the counter VC is set to 0 (step S7).
8) End this routine. The counter VC is
This is for measuring the rotation speed of the rotary weighing device 10 and is not used in this embodiment, but is used in a third embodiment described later.

Next, when an interrupt occurs, the value of the counter VC is incremented by one through steps S6, 62 and 64 (step S). Hereinafter, steps S26a, 28a, 30a, 32a, and 34a are executed.
This corresponds to steps S26, S28, S30, S32, S34 in the figure, and waits for q pulse signals Ta after closing the analog switch as shown in FIG. . When this is completed and an interrupt occurs, step S6,
After 62, 64, 80, 26a, steps S36a, 38a, 40a, 42
a, 44a and 46a are executed, which also correspond to steps S36, 38, 40, 42, 44 and 46 shown in FIG. 3, and as shown in FIG. After waiting for time, m digital weighing signals Wi from a certain weighing means
(A stored value of WINT) and a value of a register N indicating which measuring means is used. Such calculation is performed during the period when the dynamic weighing command is given. If the answer to step S66 is NO, the flag TPF
Is set to 0 (step S82), and then step S80 is executed.

When such a calculation is performed, when the rotary weighing device 10 rotates without the article 20 being loaded on each weighing unit and the initial centrifugal force error component storage command is given from the keyboard 44, the sixth As shown in FIG. 7A, it is determined whether the flag WEGF is 1 (step S84). If this answer is NO, this routine ends. If the answer to step S84 is YES, the stored value of the register WINT is divided by m, and the total value W of the initial load W _I of a certain weighing means, the zero point variation W _z and the initial centrifugal force error ΔW _I is obtained. _{Find GSN} (step S86),
The flag WEGF is set to 0 (step S88). Then, according to the value of the register N, the initial load W _{IN of the} weighing means,
It reads and zeros variation W _ZN, subtracts them from the W _GSN,
The absolute value [Delta] W _IN the determined (step S90), which was accumulated and stored in the accumulating register ShigumaderutaW _IN (step S9
2). Then, the number of counters CAC for counting the number of weighing means for which the calculation of the initial centrifugal force error has been completed is advanced by one (step S94), and it is determined whether or not the count value is equal to the number n of all weighing means (step S96). If the answer is no, this routine ends. Hereinafter, when the accumulated value of the initial centrifugal force error in the entire weighing means is stored in the register ShigumaderutaW _IN similarly, the value of the counter CAC is n, and the answer of step S96 YE
Become S. At this time, the value of the register ShigumaderutaW _IN is the cumulative value of the initial centrifugal force error in the entire weighing unit divided by n, stores the average initial centrifugal force error [Delta] W _I of the metering means (Step S9
8) The counter CAC is set to 0 (step S100), and this routine ends. In this routine, the average value of the initial centrifugal force error of each weighing means is obtained, and the average value is used as the initial centrifugal force error of each weighing means. Because there is. When it is necessary to obtain the initial centrifugal force error of each measuring means with higher accuracy, the initial centrifugal force error may be measured a plurality of times for each measuring means and averaged.

If an article having a reference weight M is loaded on each of the weighing means and the centrifugal force error component storage command at the time of loading the reference article is given from the keyboard 44 in a state where the rotary weighing apparatus 10 is operated, the sixth command is issued. As shown in FIG. 6B, it is determined whether the flag WEGF is 1 (step S100). If the answer is NO,
This routine ends. If the answer to step S100 is YES, the stored value of the register WINT is divided by m, and the initial load W _I , the zero-point fluctuation W _Z , the reference weight M, and the centrifugal force error ΔW _XN when a certain article is loaded are obtained. the average value W _GSN those containing calculated (step S102), the flag WEGF to 0 (step S1
04). Then, the initial load W _IN specified by the register N
It is determined whether the value obtained by subtracting the zero point variation _WZN from the _WGSN is smaller than the reference weight M and greater than a predetermined threshold M1 (step S106). If this answer is NO, the routine is terminated because it has been forgotten to load the reference article on the weighing means. If the answer to step S106 is YES, since the reference article is loaded, the reference weight M and the register N
Initial load W _IN and the zero-point variation W _ZN subtracted from W _GSN, reference article loading Centrifugal force error ΔW its absolute value but is specified
_XN is obtained (step S108), and this is accumulated and stored in the accumulation register ΣΔW _XN (step S110). Then, the value of the counter CAC that counts the number of weighing means that has finished calculating the centrifugal force error when the reference article is loaded is incremented by 1 (step S112), and it is determined whether the value of the counter CAC is equal to n (step S114). . If the answer is NO, this routine is terminated, and the above-described processing is performed every time the flag WEGF is set to 1. And the answer of step S114 is
When YES, the register ΣΔW _XN stores the n accumulated values of the centrifugal force error at the time of loading the reference article, and divides this by m to obtain the average value of the centrifugal force error at the time of loading the reference article Δ
W _X is obtained and stored (step S116), and the counter
The CAC is set to 0 (step S118), and this routine ends. If the reference article loading centrifugal force error is to be calculated with higher accuracy, a plurality of reference article loading centrifugal force errors may be individually measured for each weighing means, and the average value thereof may be obtained. Good.

When the dynamic normal weighing command is given from the keyboard 44, the weight of each article 20 in a state where each weighing means is rotating is performed as shown in FIG. That is, it is determined whether the flag WEGF is 1 (step S120).
If NO, this routine ends. If this judgment is YES, the stored value of the register WINT is divided by m,
The average value W _GSN of the initial load W _IN of a certain weighing means, the zero-point variation W _ZN , the weight of the article 20 and the centrifugal force error during measurement is calculated (step S122), and the flag WEGF is set to 0 (step S12).
Four). Then, the initial load W _IN and the zero-point variation W _ZN corresponding to the weighing means specified by the register N are subtracted from W _GSN to obtain W _dmN (step S126), and whether W _dmN is larger than the threshold value M1 is determined. A determination is made (step S128). If this answer is YES, since the article 20 is loaded on this weighing means, the initial load W _IN and the zero-point variation W _ZN are subtracted from W _GSN , and this is added to the centrifugal force error when the reference article is loaded. Add ΔW _X ,
The true weight W _dMN of the article 20 is calculated (Step S130). Then, using this weight W _dMN , a process is performed to determine whether the article 20 is non-defective or defective (step S132), and this routine ends.

Further, if it is the answer to step S128 is NO, the metering means so not loading the article 20, W _GSN is intended to include the initial load W _IN and zeros variation W _ZN and the initial centrifugal force error [Delta] W _I. Therefore, the initial load W _IN is subtracted from W _GSN, by adding the initial centrifugal force error [Delta] W _I to obtain the latest zero-point variation W _ZN of the weighing means (step S134), it registers N is This is stored in the designated zero point register (step S136), and this routine ends. Therefore, zero point variation W _ZN subsequent step S126, the latest zero-point variation W _ZN are used.

The first embodiment measures the initial centrifugal force error and the centrifugal force error when the reference article is loaded at the place of use of the rotary weighing device 10 and stores them. These are stored in advance in a factory or the like. That is, the initial centrifugal force error ΔW _I is calculated by ΔW _I = m · V ² · tan θ / r. m is the initial load, taking a constant value, tan θ
Is also a constant value because m is a constant value, and r is also a constant value. Therefore, ΔW _I is expressed as ΔW _I = k · V ² . Here, k is a proportional constant. Generally, the speed value used by the worker is the moving distance V per unit time (m / m
Since the rotation speed per unit time is R (r · p · m) rather than s), ΔW _I can be expressed as ΔW _I = K · R ² . Here, K is a proportional constant. Therefore, the proportionality constant K is K = ΔW _I / R ² . The rotary centrifugal weighing device 10 is rotated at a predetermined rotation speed R, and the initial centrifugal force error ΔW _{I is} determined in the same manner as in the first embodiment.
Is measured, and a proportional constant K is obtained based on the above equation. Do this for various R, in correspondence with R, may be stored by setting the R, it reads the K corresponding thereto, calculates the initial load centrifugal force error [Delta] W _I. Alternatively, the initial load centrifugal force error ΔW _I is stored corresponding to each R, and the initial load centrifugal force error Δ
W _I may be read. Further, in correspondence with the R to store the code and [Delta] W _I, it may be read out [Delta] W _I or K by setting the code.

The same applies to the case of the centrifugal force error ΔW _X when the reference article is loaded.
Since ΔW _X is expressed as ΔW _X = k1 (W _I + M) ² · V ² = K 1 (W _I + M) ² · R ² , a reference weight M and a rotation speed R are set.
As in the first embodiment, obtains the centrifugal force error [Delta] W _X when the reference article loading, obtains the _{K1 = (W I + M)} 2 · R 2 / ΔW X as K1. This is obtained for various R and M, stored, R and M are set at the time of use, K1 is read, and ΔW _X is calculated based on K1. Alternatively, instead of storing K1, various ΔW _Xs may be stored in association with R and M, and by setting R and M, the corresponding ΔW _X may be read out. It should be noted that a code may be attached to each of various combinations of R and M, and ΔW _X may be read by setting this code. This eliminates the need for the user to measure and store the initial load centrifugal force error and the reference article loading error. Moreover, the centrifugal force error at the time of loading the reference weight and the centrifugal force error at the initial load can be conveniently read out according to the rotational speed R that is normally used. In the first and second embodiments, the reference article having the reference weight M is used, and the centrifugal force error ΔW when the reference article is loaded is used.
_{Although XN} was determined, the reference article does not necessarily have to have the reference weight M. For example, an article considered appropriate from a group of articles to be weighed is selected and the
Is stopped, the weight is loaded on one of the weighing means, the static weight M 'is measured, this M' is stored, and then the rotary weighing device 10 is rotated to obtain the _WGSN . , And M ′ to obtain ΔW _XN . However, since the sample is one, to improve the reliability, several sought [Delta] W _XN performed as described above a plurality of times, it is to the average value and the formal [Delta] W _XN.

In the third embodiment, a centrifugal force error when an article is loaded is calculated based on the rotation speed of the rotary weighing device 10 and the weighed value at that time. That is, the centrifugal force error Δ when the article is loaded
W _XI , ΔW _XI = k ² · W ² / C ² . Where W is the weight of the article W _X and the initial load of the weighing means
Sum of the W _I, C is a value metering device is inversely proportional number of pulse signals occurring during the movement of the predetermined distance, i.e., the velocity V,
k2 is a proportionality constant. Therefore, the above equation can be expressed as ΔW _XI = k2 (W _X + W _I ) ² / C ² . W _I is capable of measuring and storing in the same manner as in the first embodiment, C is a pulse signal T used in the first embodiment
It can be measured using P or the pulse signal RP. However, W _X cannot be measured. The weighing signal ΔW at this time
_GSN, in addition to the W _X and W _I, but contains a zero-point variation W _Z and [Delta] W _XI, contains in addition to [Delta] W _XI of future minus the W _I and W _Z, i.e. W _X Even if ΔW _XI is obtained by using the value in place of W _X , a large error does not occur because ΔW _XI itself is not a very large value. Therefore, To obtain ΔW _XI . Note that when measuring with the metering means to W _I, the zero drift and the like of deviation and measurement circuit of the zero point of the load cell, the measured value may not be correct. Therefore, the part corresponding to the initial load is measured by another measuring means or calculated logically, and the initial load _WIA
And ΔW _XI , It is possible to obtain ΔW _XI with higher accuracy.
In this way, when determining the [Delta] W _XI, it obtains the weight W _X of the article 20 by _{W X = W GSN -W I -W} Z + ΔW XI.

To determine ΔW _XI as described above, k2 must be determined. K2 may be determined by numerical calculation, but the most practical method may be as follows.

That is, as in the first embodiment, obtaining the initial load centrifugal force error [Delta] W _I. At the same time, by counting the TP pulse signal or RP pulse signal to determine the count number C _a which is proportional to speed. Since the initial load centrifugal force error ΔW _{I at} this time is ΔW _I = k ² · W _IA ² / C _a ² , k ² is obtained by k 2 = ΔW _I · C _a ² / W _IA ² .

The flowchart for obtaining k2 in this manner is shown in FIG.
Figure 10 shows. Also in this case, the interrupt routine shown in FIGS. 2 and 3 is performed every time the pulse signal Ta is generated. And do not load the goods on each weighing means,
Assuming that the rotation is performed, every time the flag WEGF becomes 1, m accumulated values of the weighing values including the initial load of each weighing means, the zero point variation, and the initial load centrifugal force error are obtained in the register WINT. Therefore, first, it is determined whether the flag WEGF is 1 (step S150), and if the answer is NO, this routine is ended. If the answer to step S150 is YES, the value of WEGF is divided by m, and m average values W of the weighing values including the initial load, the zero-point variation, and the initial load centrifugal force error of each weighing means are obtained.
_{The GSN} is obtained (step S152), and the flag WEGF is set to 0 (step S154). Then, the value of the register CAU is set to C, which is accumulated and stored in the accumulation register $ C (step S156). Here, the value of the register CAU is obtained by transferring the value of the counter VC as shown in step S78 of FIG. 2, and this VC is obtained by counting the pulse signal Ta as shown in step S80. This operation is performed during the period from when one analog switch is closed to when the next analog switch is closed. That is, the time from when one measuring means reaches the position a shown in FIG. 7 to when the next measuring means reaches the position a is measured. That is, it is proportional to the speed of the weighing means.

Following step S156, the initial load W _I, the sum of the zero-point variation W _Z of the metering means, and subtracted from W _GSN, stores the absolute value as an initial centrifugal force error [Delta] W _IN (step S15
8). Then, this is accumulated, and the accumulated value is stored in the register Σ
The initial centrifugal force error ΔW is stored in ΔW _IN (step S160).
Counter CAC that counts the number of weighing means that measured _IN
Is incremented by 1 (step S162), and it is determined whether the value of the counter CAC is equal to the number n of all the weighing means (step S16).
Four). If the answer is no, this routine ends. Hereinafter, similarly to the initial centrifugal force error [Delta] W _IN of the metering means is accumulated, when the number is n, the answer to step S164 is YES
Then, the stored value of the register ΣΔW _IN is divided by n to obtain an initial centrifugal force error ΔW _I, and the value of the register ΣC is
To obtain an average speed Ca (step S166). Then, the calculation of ΔW _I · C _a ² / W _IA ² is performed to obtain k2 (step S168), and the counter CAC is set to 0 (step S17).
0), end this routine.

After setting k2 in this way, the article 20
Is loaded, the processing shown in the flowchart of FIG. 11 is performed. In this case as well, the interrupt routine shown in FIGS. 2 and 3 is executed, and every time the flag WEGF becomes 1, a value obtained by accumulating m weighing values of the weighing means reaching the point a is stored in the register WINT. It is memorized. Therefore, first, it is determined whether the flag WEGF is 1 (step S172), and if the answer is NO, this routine ends. If the answer to step S172 is YES, the value of the register WINT is divided by m, and the average value _WGSN of the sum of the weight of the article 20, the initial load, and the zero-point variation minus the centrifugal force error is calculated. Then (step S174), the flag WEGF is set to 0 (step S176). Then, the value of the register CAU is set to C (step S178). This represents the speed of the metering means at point a. And k2 ｛W _IAN + W _GSN −
_Perform the calculation of (W _IN + W _ZN )｝ / C ² to obtain the centrifugal force error ΔW _IX
Is calculated (step S180). And W _IN and W from W _GSN
Subtracted and _ZN, this was added to the centrifugal force error [Delta] W _IX, obtains the metering true value W _dN from the metering means (step S182). Then, W _dN threshold W _a greater or determined (step S18
4) If the answer is YES, since the article on the weighing means is loading, with W _dN, performs sorting processing (step S186), and terminates this routine.

On the other hand, when the answer to step S184 is NO, the on metering means, since the article 20 is not loading, W _GSN -W _dN -W _IN +
The calculation of ΔW _IX is performed to obtain a new zero-point variation W _ZN (step S188), which is stored in the zero-point register designated by the counter N (step S190), and this routine ends.

[Effects of the Invention] As described above, according to the present invention, a centrifugal force error can be corrected, and measurement can be performed with extremely high accuracy.
In particular, according to the first aspect, the fluctuation of the zero point can be corrected at any time, so that the measurement can be performed with high accuracy. Further, according to the first invention, only an article having a weight near a specific reference weight can be measured with high accuracy, but according to the second invention, even if the article has an arbitrary weight, the centrifugal force error can be corrected. Can be. Further, according to the third aspect, the centrifugal force error can be automatically corrected by setting only the rotation speed and the reference weight.

[Brief description of the drawings]

FIG. 1 is a flowchart of the dynamic normal weighing of the first embodiment of the rotary weighing device according to the present invention, FIG. 2 is a flowchart of a part of an interrupt routine of the first embodiment, and FIG. The remaining flowchart of the interrupt routine of the first embodiment, FIG. 4 is a flowchart of the first embodiment when the power is turned on, FIG. 5 is a flowchart of the stationary state of the first embodiment, and FIG. FIG. 7 is a flowchart at the time of rotation of the first embodiment, FIG. 7 is a mechanical configuration diagram of the first embodiment, FIG. 8 is a block diagram of the first embodiment, and FIG. FIG. 10 is a flowchart for determining the proportionality constant of the third embodiment, FIG. 11 is a flowchart for the dynamic weighing of the third embodiment, and FIG. 12 is an explanation of the centrifugal force error. FIG. 10 ... rotary measuring device, 16 _{1 to} 16 _n ... measuring means, 38 ...
…CPU.

Claims (3)

(57) [Claims] 1. A rotary weighing device having a weighing device that rotates at a constant speed around a predetermined center of rotation, wherein the rotary weighing device is in an initial state in which no articles are loaded on the weighing device, and the rotary weighing device is in a non-rotating state. Measuring the static variation that is the output of the weighing means when the weighing means is in the initial state and the output of the weighing means when the rotary weighing device is rotating at a predetermined speed. Measuring an initial centrifugal force error, which is a vertical component of a force generated to balance a centrifugal force generated based on rotation in the initial state of the rotary weighing device based on the stationary fluctuation component; The rotary weighing device is based on the output of the weighing device and the stationary variation when the rotary weighing device is rotating at the predetermined speed while the reference article is loaded on the weighing device. Above Obtaining a reference article loading centrifugal force error, which is a vertical component of the force generated to balance the centrifugal force generated based on the rotation in the reference article loaded state, and at the predetermined speed of the rotary weighing device. Determining whether or not an article to be weighed having a weight close to the weight of the reference article is loaded in the rotating state, and an output of the weighing means when it is determined that the article to be weighed is loaded. Calculating the weight of the article to be weighed based on the stationary variation and the centrifugal force error when the reference article is loaded, and the output of the weighing means when it is determined that the article to be weighed is not loaded. Correcting the static variation based on the initial centrifugal force error,
Rotary weighing method equipped. 2. A rotary weighing device having a weighing means that rotates at a constant speed around a predetermined center of rotation, wherein the weighing means is in an initial state where no articles are loaded and the rotary weighing apparatus is in a non-rotational state. Measuring the initial value variation which is the output of the weighing means when the weighing means is at a position; determining the true static weighing value of the weighing means itself; and Based on the output of the weighing device in the measurement state where the weighing device is rotating, the rotation speed of the rotary weighing device, the initial value variation, and the true static weighing value of the weighing device, Calculating a centrifugal force error which is a vertical component of a force generated to balance a centrifugal force generated during rotation in the loaded state of the article to be weighed; Calculating the weight of the article to be weighed based on the centrifugal force error. 3. A rotary weighing device having a weighing device that rotates at a constant speed around a predetermined center of rotation, wherein the rotary weighing device is in an initial state in which no articles are loaded on the weighing device and the rotary weighing device is in a non-rotating state. Measuring the static variation, which is the output of the weighing means when the weighing means is in the position, and the reference at the rotational speed based on the weight of the reference article, the rotational speed of the rotary weighing device, and the initial load of the weighing means. Calculating a reference article loaded centrifugal force error which is a vertical component of a force generated to balance the centrifugal force generated based on the rotation in a state where the article is loaded on the weighing means; and When an article to be weighed having a weight close to the weight of the reference article is loaded and the rotary weighing apparatus is in a rotating state, the output of the weighing means, the stationary fluctuation, and the centrifugal force error when the reference article is loaded. Calculating the weight of the article to be weighed based on the above.